Agent skill

Ceos L10

by bradfeld in bradfeld/ceos

A skill your agent uses when running or reviewing a Level 10 weekly leadership meeting

MITAuto-check passed

Install Ceos L10

skills CLI
$ npx skills add bradfeld/ceos --skill ceos-l10 -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install bradfeld/ceos ceos-l10 --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/bradfeld/ceos.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/ceos-l10 .claude/skills/ceos-l10 && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
ceos-l10
GitHub stars
156
Token cost
~3.3k tokens
SKILL.md length
1,856 words
Files
1
Skills in repo
23
Repo updated
First seen
Licence
MIT

At a glance

A skill your agent uses when running or reviewing a Level 10 weekly leadership meeting

  • Works in 9 steps: Set Up the Meeting → Segue (5 minutes) → Scorecard Review (5 minutes) → …
  • Reviewing a Level 10 weekly leadership meeting
  • SKILL.md covers When to Use, Context, Process and Transcript Mode, plus 3 more sections
  • Calls git

What it does

Ceos L10 is an agent skill from bradfeld/ceos. Use when running or reviewing a Level 10 weekly leadership meeting

Its SKILL.md is about 3.3k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

The repository describes itself as: EOS (Entrepreneurial Operating System) skills for Claude Code. Run your business on EOS with AI. The licence is MIT.

When your agent uses it

  • Reviewing a Level 10 weekly leadership meeting

Example prompts

  • “/ceos-l10”

Workflow steps

9 steps, taken from the step headings in SKILL.md.

  1. Set Up the Meeting
  2. Segue (5 minutes)
  3. Scorecard Review (5 minutes)
  4. Rock Review (5 minutes)
  5. Headlines (5 minutes)
  6. To-Do Review (5 minutes)
  7. IDS (60 minutes)
  8. Conclude (5 minutes)
  9. Save the Meeting

What it can do on your machine

Read from SKILL.md and the folder at commit bdd5f01. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • git

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use git, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Ceos L10 loads about 3.3k tokens when it runs. Until then it costs about 19 tokens; SKILL.md has 1,856 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~19
When it runs · the whole SKILL.md, loaded when a task matches
~3.3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from bradfeld/ceos at commit bdd5f01, republished under its MIT licence (© bradfeld). 1,856 words, ~3,316 tokens.

Download SKILL.mdSave it as .claude/skills/ceos-l10/SKILL.md (or your agent's skills folder).
name
ceos-l10
description
Use when running or reviewing a Level 10 weekly leadership meeting
file-access
data/meetings/l10/, templates/l10-meeting.md, data/scorecard/weeks/, data/rocks/, data/issues/open/, data/accountability.md
tools-used
Read, Write, Glob

ceos-l10

Facilitate the Level 10 (L10) Meeting — the weekly 90-minute leadership meeting that keeps the team aligned, accountable, and solving problems. The L10 follows a strict 7-section agenda with time boxes.

When to Use

  • "Run L10" or "start our L10"
  • "Level 10 meeting" or "weekly meeting"
  • "Start our meeting" or "let's do our weekly"
  • "Review last week's L10" or "what were our to-dos?"
  • Any weekly leadership team meeting

Context

Finding the CEOS Repository

Search upward from the current directory for the .ceos marker file. This file marks the root of the CEOS repository.

If .ceos is not found, stop and tell the user: "Not in a CEOS repository. Clone your CEOS repo and run setup.sh first."

Sync before use: Once you find the CEOS root, run git -C <ceos_root> pull --ff-only --quiet 2>/dev/null to get the latest data from teammates. If it fails (conflict or offline), continue silently with local data.

Key Files
FilePurpose
data/meetings/l10/YYYY-MM-DD.mdL10 meeting notes
data/scorecard/weeks/Latest scorecard (pulled during Scorecard section)
data/rocks/[quarter]/Current Rocks (pulled during Rock Review)
data/issues/open/Open issues (surfaced during IDS section)
data/meetings/l10/Previous L10s (for To-Do review)
templates/l10-meeting.mdMeeting template
data/accountability.mdAccountability Chart (seat owners for To-Do assignment validation)
Meeting Structure (90 minutes)
#SectionTimePurpose
1Segue5 minPersonal/professional good news
2Scorecard5 minReview weekly numbers
3Rock Review5 minStatus of quarterly Rocks
4Headlines5 minCustomer and employee news
5To-Do Review5 minCheck last week's To-Dos
6IDS60 minIdentify, Discuss, Solve top issues
7Conclude5 minNew To-Dos, cascading messages, rating

Process

Step 1: Set Up the Meeting
  1. Date: Use today's date, or ask if scheduling for a different day.
  2. Attendees: Ask who's attending. Default to the attendees from the most recent L10 if available.
  3. Create the meeting file from templates/l10-meeting.md at data/meetings/l10/YYYY-MM-DD.md.

If an L10 already exists for today, ask: "There's already an L10 for today. Open it, or create a new one?"

Step 2: Segue (5 minutes)

Prompt each attendee to share one personal and one professional good news. This builds connection before diving into business.

Record responses in the Segue section.

"Let's start with a segue. [Name], share one personal and one professional good news."

Step 3: Scorecard Review (5 minutes)
  1. Read the most recent weekly scorecard from data/scorecard/weeks/.
  2. Display the metrics table with status.
  3. For each off-track metric, note it for the Issues list.
Scorecard — Week 07:
  Weekly Revenue: \$52K (goal: \$50K) ✓
  New Customers: 7 (goal: 10) ✗ → add to Issues?
  NPS: 72 (goal: 70) ✓

Ask: "Any off-track items to add to today's Issues list?"

If no scorecard exists for this week: "No scorecard logged for this week yet. Skip, or enter numbers now?"

Step 4: Rock Review (5 minutes)
  1. Read all Rocks in data/rocks/[current-quarter]/.
  2. Display status for each Rock.
  3. Quick check — on track or off track? No discussion yet (save that for IDS).
Rock Review — Q1 2026:
  Launch Beta (brad): on_track
  Hire VP Sales (daniel): off_track → add to Issues?
  Implement CRM (daniel): on_track

Ask: "Any off-track Rocks to add to today's Issues list?"

Step 5: Headlines (5 minutes)

Prompt for customer and employee headlines — good news, bad news, updates. Quick hits only. No discussion.

"What headlines do we need to share? Customer news, employee news, anything the team should know."

Record as bullet points.

Step 6: To-Do Review (5 minutes)
  1. Read the most recent previous L10 from data/meetings/l10/.
  2. Find the "New To-Dos" table from the Conclude section.
  3. For each To-Do, ask: "Done or not done?"
To-Do Review (from L10 2026-02-06):
  ☐ Send proposal to Acme (brad) — Done? [yes/no]
  ☐ Schedule interviews (daniel) — Done? [yes/no]
  1. Calculate completion rate. Target is 90%+.
  2. If below 90%, note it but don't dwell — it's data, not a guilt trip.
Step 7: IDS (60 minutes)

This is the core of the L10 — where the real work happens.

Step 7a: Build the Issues List

Collect issues from:

  • Off-track Scorecard items (from Step 3)
  • Off-track Rocks (from Step 4)
  • Open issues from data/issues/open/
  • New issues raised by attendees

Ask: "What other issues should we add to today's list?"

Step 7b: Prioritize

List all issues and ask the team to pick the top 3 most important ones. These get solved today. The rest stay on the list for next week.

Step 7c: IDS Each Issue

For each of the top 3 issues, follow the IDS process:

  1. Identify — What is the real issue? Dig past symptoms. Use the 5 Whys if the root cause isn't obvious.
  2. Discuss — Each person shares their perspective. Stay focused — no tangents, no solving yet.
  3. Solve — What's the action? Create at least one To-Do with an owner and due date. Cross-reference data/accountability.md to validate that the To-Do owner matches the seat responsible for that area.

Record the IDS work in the meeting notes. If this creates a new issue file, use ceos-ids to write it — ceos-l10 reads data/issues/open/ but does not write there directly.

Step 8: Conclude (5 minutes)
New To-Dos

Compile all To-Dos created during the meeting (from IDS and any other sections):

To-DoOwnerDue Date
[Action][Name][YYYY-MM-DD]
Cascading Messages

Ask: "What needs to be communicated to the rest of the organization?"

Meeting Rating

Each attendee rates the meeting 1-10. Record ratings and calculate average. Target: 8+.

If average is below 8: "Below the 8/10 target. What could we improve for next week?"

Step 9: Save the Meeting
  1. Show the completed meeting file.
  2. Ask for any final edits.
  3. Write to data/meetings/l10/YYYY-MM-DD.md.
  4. Update the frontmatter with the average rating.
  5. Remind: "Run git commit to save the meeting notes."

Transcript Mode

A hybrid workflow designed for teams who want to stay off screens during the meeting and use Apple Voice Memos (or any voice transcription) to feed Claude at two natural break points.

When to Use
  • "Here's my transcript" or "fill this in from my recording"
  • "Transcript mode" or "hybrid L10"
  • Any time a voice memo transcript is pasted instead of running the meeting live
How It Works

The meeting runs in two recorded segments with Claude engaged at each break.

Break 1 — Before IDS

Stop the recording after To-Do Review. Paste the transcript and tell Claude:

  • Who was in the meeting (names, since Voice Memos doesn't label speakers)
  • That this covers sections 1–5

Claude will:

  1. Parse the transcript and fill sections 1–5 of the meeting file
  2. Build the Issues list from off-track Scorecard items, off-track Rocks, open issues in data/issues/open/, and anything raised in the conversation
  3. Suggest top 3 prioritization based on what was discussed and existing issue data
  4. Flag any gaps (missing scorecard numbers, unclear To-Do status) before IDS starts

The team walks into IDS with a ranked Issues list already prepared.

Show full SKILL.md (784 more words)Show less
Break 2 — Before Conclude

Stop the recording after IDS. Paste the transcript.

Claude will:

  1. Parse each issue discussed and fill the IDS section (Identify, Discuss, Solve)
  2. Extract all To-Dos with owners and due dates
  3. Flag any To-Dos missing an owner or due date
  4. Draft cascading messages based on decisions made
  5. Present the complete Conclude section ready to read back to the room

The team spends Conclude confirming the To-Do list and rating the meeting — nothing to reconstruct from memory.

Transcript Tips
  • Name the attendees upfront. Paste a quick header before the transcript: Attendees: Brad, Dana, Sam, Jordan. Voice Memos doesn't label speakers; Claude infers from context but a name list helps.
  • Don't clean up the transcript. Filler words, crosstalk, and incomplete sentences are fine — Claude works from meaning, not polish.
  • One paste per break. Don't combine both segments into one paste. The break point is where Claude's IDS prep adds the most value.
  • Implicit issues are fair game. If the same problem surfaces three times across different topics, Claude will flag it as a potential root cause issue even if it wasn't formally added to the list.
Guardrails
  • Never infer owners. If a To-Do doesn't have a clear owner in the transcript, Claude asks rather than guesses.
  • Never infer due dates. Same rule — prompt for missing dates rather than defaulting to "next week."
  • Preserve verbatim for IDS. The Discuss sub-section should use close paraphrases from the transcript, not summaries. This matters for anyone who missed the meeting.
  • Still save at the end. The complete meeting file is written once — after Break 2 and the Conclude confirmation — not incrementally.

Output Format

The meeting file follows the template structure in templates/l10-meeting.md:

  • YAML frontmatter: date, attendees, rating
  • 7 numbered sections with content from the meeting
  • To-Do table in the Conclude section

During the meeting, display each section header with its time box as you move through the agenda.

Guardrails

  • Respect time boxes. The L10 is 90 minutes total. Don't let any section (especially IDS) expand beyond its allocation without acknowledging the trade-off.
  • IDS gets the most time. If the team is rushing through Segue/Scorecard/Rocks/Headlines/To-Dos, that's fine — bank the extra time for IDS. But don't skip any section entirely.
  • Top 3 only. During IDS, solve the top 3 issues. Don't try to solve everything — the rest go to next week's list.
  • Pull real data. Scorecard and Rock Review should use actual data from data/scorecard/weeks/ and data/rocks/. Don't ask the user to recite numbers that are already on file.
  • Reference previous L10. To-Do review must check the actual previous meeting file, not rely on memory.
  • Don't auto-invoke skills. When IDS work creates an issue worth tracking, mention that ceos-ids can create the issue file, but let the user decide. Same for scorecard logging (ceos-scorecard).
  • One meeting per day. If an L10 already exists for the date, confirm before overwriting.
  • Save at the end. Write the complete meeting file at the end, not incrementally during the meeting.
  • Sensitive data warning. On first use, remind the user: "L10 meeting notes may contain sensitive business discussions and personnel matters. Use a private repo."

Integration Notes

Scorecard (ceos-scorecard)
  • Read: ceos-l10 reads the latest weekly scorecard from data/scorecard/weeks/ during Section 2 (Scorecard Review). Each metric owner reports their number and off-track items are flagged for the Issues list.
  • Suggested flow: Off-track metrics that persist for 3+ weeks should become Issues via ceos-ids.
Rocks (ceos-rocks)
  • Read: ceos-l10 reads Rock files from data/rocks/[current-quarter]/ during Section 3 (Rock Review). Each Rock owner reports on_track or off_track status.
  • Suggested flow: Off-track Rocks should be discussed further in the IDS section.
To-Dos (ceos-todos)
  • Read/Write: ceos-l10 reads previous meeting To-Dos during Section 5 (To-Do Review) and creates new To-Dos during Section 7 (Conclude). To-Dos created during L10 use source: l10.
  • Suggested flow: Use ceos-todos Create mode for formal To-Do tracking after the meeting.
IDS (ceos-ids)
  • Read: ceos-l10 reads open issues from data/issues/open/ during Section 6 (IDS). The L10 surfaces the prioritized list; ceos-ids handles the formal Identify, Discuss, Solve work on individual issues.
  • Suggested flow: New issues surfaced during L10 should be created via ceos-ids Create mode.
Accountability Chart (ceos-accountability)
  • Read: ceos-l10 reads data/accountability.md during IDS (Step 7c) to validate that To-Do owners match seat responsibilities. When assigning action items, the person who owns the relevant seat should own the To-Do.
  • Suggested flow: If a To-Do is assigned to someone outside the responsible seat, flag it: "This action falls under the [Seat] responsibilities. Should [Seat Owner] own it instead?"
Orchestration Principle

ceos-l10 is the orchestrator — it reads data from multiple skills during the meeting but defers to each skill for formal create/update operations. The meeting file at data/meetings/l10/YYYY-MM-DD.md is the only file ceos-l10 writes directly.

© bradfeld, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in skills/ceos-l10 of bradfeld/ceos.

Open the folder on GitHubat commit bdd5f01

Compare with similar skills

Ceos L10 next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Ceos L10 compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Ceos L10 this skillbradfeld/ceos156—~3.3kAutomated safety check: PassMIT
Weekly Project Digeststhedotmack/claude-mem99k—~3.5kAutomated safety check: PassApache-2.0
Weekly Review PlanningNousResearch/hermes-agent252k—~994Automated safety check: PassMIT
Ceo Advisordavila7/claude-code-templates32k1 repos~2.8kAutomated safety check: PassMIT
CEO Plan Reviewgarrytan/gstack136k—~2.6kAutomated safety check: PassMIT
Ceo Advisoralirezarezvani/claude-skills28k1 repos~2kAutomated safety check: PassMIT

Similar skills

  • Weekly Project Digests

    thedotmack/claude-mem

    Turns a project's claude-mem timeline into a week-by-week narrative, splitting it by ISO week and running one subagent per week that receives the prior week's carry-forward block.

    99k GitHub stars~3.5k tokensUpdated yesterday
    Agent WorkflowsAuto-check passed
  • Weekly Review Planning

    NousResearch/hermes-agent

    Weekly reset: commitments, stalled work, next-week plan. An agent skill from NousResearch/hermes-agent.

    252k GitHub stars~994 tokensUpdated today
    Knowledge ManagementAuto-check passed
  • Ceo Advisor

    davila7/claude-code-templates

    Executive leadership guidance for strategic decision-making, organizational development, and stakeholder management.

    32k GitHub starsUsed in 1 repo~2.8k tokens
    Business, Finance & HRAuto-check passed
  • CEO Plan Review

    garrytan/gstack

    Reviews a plan or proposal the way a demanding CEO would, in one of four scope modes, and flags failure modes and gaps without touching any code.

    136k GitHub stars~2.6k tokensUpdated today
    Product & Project ManagementAuto-check passed
  • Ceo Advisor

    alirezarezvani/claude-skills

    Executive leadership guidance for strategic decision-making, organizational development, and stakeholder management.

    28k GitHub starsUsed in 1 repo~2k tokens
    Business, Finance & HRAuto-check passed
  • Weekly Review

    alirezarezvani/claude-skills

    A skill your agent uses when someone wants to run a weekly review, close open loops, audit stalled projects and commitments, get their system back to trusted, restart a lapsed review habit, or says…

    28k GitHub stars~1.4k tokensUpdated 1 mo ago
    Knowledge ManagementAuto-check passed

More from bradfeld/ceos

All 23 skills in this repo
  • Ceos Accountability

    bradfeld/ceos

    A skill your agent uses when viewing, updating, or auditing the Accountability Chart — seats, owners, and roles

    156 GitHub stars~3.4k tokensUpdated 26 days ago
    Auto-check passed
  • Ceos Assistance

    bradfeld/ceos

    A skill your agent uses when managing daily operational delegation through The Stack and structured leader-assistant standups

    156 GitHub stars~3.9k tokensUpdated 26 days ago
    Auto-check passed
  • Ceos Calendar

    bradfeld/ceos

    A skill your agent uses when managing the rolling market calendar - conferences, launches, fundraising milestones, partner events, and team constraints

    156 GitHub stars~2k tokensUpdated 26 days ago
    Auto-check passed
  • Ceos Cashflow

    bradfeld/ceos

    A skill your agent uses when assessing and optimizing the 8 financial levers that drive cash flow and profitability

    156 GitHub stars~3.2k tokensUpdated 26 days ago
    Auto-check passed
  • Ceos Clarity

    bradfeld/ceos

    A skill your agent uses when taking a Clarity Break — stepping back from day-to-day work for strategic thinking time

    156 GitHub stars~3k tokensUpdated 26 days ago
    Auto-check passed
  • Ceos Dashboard

    bradfeld/ceos

    A skill your agent uses when you want a quick snapshot of overall business health across all EOS components

    156 GitHub stars~3.2k tokensUpdated 26 days ago
    Auto-check passed

Questions about Ceos L10

What does Ceos L10 do?

A skill your agent uses when running or reviewing a Level 10 weekly leadership meeting. Ceos L10 is an agent skill from bradfeld/ceos.

When should I use Ceos L10?

Ceos L10 fits situations like: reviewing a Level 10 weekly leadership meeting.

How do I install Ceos L10 in Claude Code?

Run `npx skills add bradfeld/ceos --skill ceos-l10 -a claude-code`. Or copy the skill folder (skills/ceos-l10 in bradfeld/ceos) into .claude/skills/ceos-l10 in your project. Claude Code loads it when a task matches its description.

How do I install Ceos L10 in Codex?

Run `npx skills add bradfeld/ceos --skill ceos-l10 -a codex`. Or copy the skill folder (skills/ceos-l10 in bradfeld/ceos) into .agents/skills/ceos-l10 in your project. Codex loads it when a task matches its description.

Can I use Ceos L10 in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add bradfeld/ceos --skill ceos-l10 -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/ceos-l10, .gemini/skills/ceos-l10, .github/skills/ceos-l10 and .opencode/skills/ceos-l10 in your project.

What does Ceos L10 need to run?

Going by SKILL.md and its folder, Ceos L10 needs the command-line tools its instructions call (git).

Does Ceos L10 access the network?

SKILL.md contains no URLs. Its commands use git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Ceos L10 safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Ceos L10 use?

Ceos L10 is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Ceos L10 use?

About 3.3k tokens (SKILL.md is roughly 13k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Ceos L10?

Skills that share tags, products or a category with Ceos L10: Weekly Project Digests (thedotmack/claude-mem, 99k stars), Weekly Review Planning (NousResearch/hermes-agent, 252k stars), Ceo Advisor (davila7/claude-code-templates, 32k stars) and CEO Plan Review (garrytan/gstack, 136k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Ceos L10?

bradfeld (a GitHub user) maintains it in bradfeld/ceos, which has 156 GitHub stars. The repository holds 23 skills in this directory. The repository was last updated on September 13, 2026.

Source: bradfeld/ceos on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.